Sir,

We read with interest the article by Homer et al.1 We would like to make some statistical and clinical observations.

The subset of eyes included received an average of 23.8±18.8 injections of ranibizumab or bevacizumab over a period of 28±20.5 months. In view of such a large standard deviation, it would have been more accurate to mention the median for the above two values. The mean interval between aflibercept injections was noted to be 59.3 days and that between ranibizumab/bevacizumab was noted as 37±6.1 days. Without the mention of standard deviation for aflibercept, it is difficult to gauge a statistical comparison between the two groups as one cannot ascertain whether the distribution of the intervals for the aflibercept group was parametric or non-parametric. Without that information, application of the t-test may not be appropriate.

The inclusion criteria at IRB approval mentions cases with ≥6 prior intravitreal ranibizumab/bevacizumab injections. But further in the article, the range of prior injections given mentions 4–62 injections. Whether the pathology can be termed as recalcitrant enough to change the intravitreal agent after just 4 injections is little questionable as past literature shows fairly good long-term response to PRN basis injection of ranibizumab over 1–2 years.2, 3 The treatment protocol for aflibercept followed in this study seems to be 2-monthly injections after the three loading dose injections as authors mention a treatment at a fixed interval of 8 weeks. With that perspective, the fact that the minimum exudation-free period noted was 35 days, it is unclear whether any patient was given repeat aflibercept before 8 weeks.

We would appreciate if the authors could clear the above doubts.